BrierCurves: Plotting Brier Curves
In paumoal/costcurvespkg: Display and analyze Cost Curves and treshold choice

Description Usage Arguments Details Value Author(s) References See Also Examples

Function to plot loss against operating condition using the score-driven threshold choice method (Brier Curves)

BrierCurves(predictions,classes,uniquec=FALSE, loss2skew=FALSE, hold=FALSE,
            plotOFF=FALSE, gridOFF=TRUE, pointsOFF=TRUE, legendOFF=FALSE,
            main, xlab, ylab, namesClassifiers, lwd, lty, col, pch, cex,
            xPosLegend,yPosLegend, cexL)

`predictions`	A list of predictions arrays, each array contains predicted scores of a specific classifier.
`classes`	A list of classes arrays, each array contains binary classes.
`uniquec`	If it is TRUE, the same array of classes is used for each array in a list predictions.
`loss2skew`	If it is TRUE, loss is computed with respect to Skew otherwise loss is used.
`hold`	If it is TRUE, the plot is not closed. This is useful to include new curves above the current curve.
`plotOFF`	Disable/enable plot visualization, only return AUC values.
`gridOFF`	Disable/enable grid visualization.
`pointsOFF`	Disable/enable point marks visualization.
`legendOFF`	Disable/enable legend visualization.
`main`	Title of the plot.
`xlab`	x label.
`ylab`	y label.
`namesClassifiers`	An array with names of each classifier.
`lwd`	Line width.
`lty`	Line type.
`col`	Line color.
`pch`	Point type.
`cex`	Size point.
`xPosLegend`	x coordinate to be used in the position of the legend.
`yPosLegend`	y coordinate to be used in the position of the legend.
`cexL`	size of box legend.

Definition:

Function that plots the expected cost/skew against loss. For a given probabilistic classifier and operating condition defined by cost proportion, the Score-Driven threshold choice method sets the threshold equal to the operating condition (cost proportion or skew).

The Brier curve for a given classifier is defined as a plot of loss against operating condition using score-driven threshold choice method.

Assuming the score-driven threshold choice method, expected loss under a uniform distribution of cost proportions is equal to the Brier score. Using skews, we arrive at the prior-independent version of the Brier score.

An array with AUBC (Area Under Brier Curve) for each test.

Paulina Morillo: paumoal@inf.upv.es

Ferri, C., Hernandez-orallo, J., & Flach, P. A. (2011). Brier curves: a new cost-based visualisation of classifier performance. In Proceedings of the 28th International Conference on Machine Learning (ICML-11) (pp. 585-592).

CostCurves, KendallCurves, predictions, RateDrivenCurves, CostLines, TestOptimal, TP_FP.rates, TrainOptimal

#Load data
data(predictions)

#Loss by cost
R<-BrierCurves(list(predictions$A, predictions$B), list(predictions$classes),
	uniquec = TRUE, loss2skew = FALSE)


#Loss by skew
R<-BrierCurves(list(predictions$A, predictions$B), list((1-predictions$classes),
	predictions$classes), loss2skew = TRUE, gridOFF = FALSE, main=NULL)